JP4187719B2 - Multi-channel downmixing equipment - Google Patents

Description

Inventor:
David H. Glee Singer (David H. Griesinger)
Background of the Invention TECHNICAL FIELD This application is filed with David H. et al., Filed May 3,2002. Claims priority to US Provisional Application No. 60 / 377,661 entitled "Multi-Channel to Two-Channel Mixing Apparatus and Method" by Glee Singer, Attorney Docket No. 11336/137. The provisional application is incorporated herein by reference.
2. TECHNICAL FIELD The present invention relates to a mixing apparatus, and more specifically, an output including a plurality of channels by mixing a multichannel signal including a plurality of channels while maintaining an intended direction and signal energy of the multichannel signal. The present invention relates to a downmixer that can be used as a signal.

  Often, an audio recording or movie soundtrack (film mix) is generated with more than two audio channels to give the listener a more realistic sense that the audio recording is live. For example, a film mix can be generated as a three-channel recording that provides a left front (LF) channel, a right front (RF) channel, and a center (C) channel. Instead, as a five-channel recording that includes a rear left (RL) channel and a rear right (RR) channel along with an LF channel, an RF channel, and a C channel, or in some circumstances, this five-channel recording channel and a lower frequency The film mix may be generated as a 5.1 channel recording that includes (LFE) channels.

  However, an audio recording or film mix listener may have an audio system that supports fewer channels than the number of channels on which the audio recording or film mix was generated. This typically occurs when the listener's audio system supports only two-channel (ie, stereo) playback. In this case, such a recording is provided to the listener as a two-channel recording by using a combiner (downmixer) that combines or downmixes multichannel signals into two channels. Downmixing may be performed, for example, with an encoder in which two-channel recording is provided on the medium (ie, CD, DVD, etc.). Downmixing may be performed at the listener of the listener's audio system, in which case the decoder downmixes the multichannel signal into a two-channel mix.

  When downmixing a multichannel signal into two channels, the downmixer typically uses a fixed mixing factor. A typical downmixer used for 5-channel film recording mixes the two rear channels before mixing the two rear channels in antiphase with the output channel. This allows the signal in the back channel to be reproduced from the back in a standard film decoder. However, information about whether the sound is from the left rear or the right rear is usually lost.

  For example, a typical downmixer for classical music that uses the European standard for 5-channel downmixing mixes the two back channels directly into the output channel without inverting the phase. This can preserve the left-right direction of the back channel, but does not preserve the indication that the signal was intended to be heard from behind the listener. The resulting mixing causes the downmixed signal to appear as if it were from the front of the listener, both during two-channel playback and when played through a standard film decoder. .

  Some downmixers can vary the mixing ratio slightly, for example, in an attempt to maintain signal energy when the peripheral input signals are anti-correlated with each other. However, for example, when the input signal pans between a plurality of input channels, the signal energy of the multi-channel signal and the apparent direction are not substantially maintained.

  Furthermore, both standard film downmixers and European standard downmixers attenuate the back channel by 3 dB before mixing the back channel into the output channel. This attenuation can cause the magnitude of the acoustic effect applied to one of the rear channels to be lower than the original 5-channel mixing. In this case, the post-input energy is not maintained in the output channel.

  Another problem with the above encoder / decoder is that there is a clear beginning with a clear beginning, such as a sound from a musical instrument or a syllable of a word, when the input signal is downmixed. Short bursts of sound with or without). The downmixing algorithm used reduces the emphasis of acoustic events in the downmixed signal, especially when there is reverberation. The downmixer downmixes the acoustic event in the previous channel. However, these acoustic events can be difficult or even inaudible when downmixed to the previous channel.

  Furthermore, the downmixer that mixes the three previous channels into the two output channels has the problem of direction localization. That is, the sound mixed in the three-channel recording so that it is perceived as coming from the middle of the left (or right) front channel and the center channel, the three-channel signal is downmixed into two channels and two When played through a speaker, it is perceived as coming from a different location. In practice, the acoustic image in two-channel downmixing is almost exactly at the left speaker (or right) rather than exactly in the middle and left.

  Therefore, there is a need for a downmixer that maintains the intended direction and signal energy of the multichannel mix. Furthermore, there is a need for a downmixer that properly mixes the input signal when reverberation is present and emphasizes the acoustic events in the input signal during the downmixing process.

SUMMARY A downmixer system is provided that generates mixing coefficients for downmixing a multichannel input signal having multiple input channels into an output signal having multiple output channels. Mix according to the comparison of the energy between the downmixed (output) signal and the input signal to the downmixer so that the energy and the desired direction of the input signal are substantially maintained in the output signal Coefficient can be made. The number of input channels of the input signal may be greater than or equal to the number of output channels of the output signal. Additionally or alternatively, the generation of mixing coefficients can maintain, for example, the intended direction of the input signal received on the peripheral input channel in at least one output channel of the output signal. In this case, it is possible to put the peripheral channel information in the peripheral channel of the upmix by using the intended direction maintained by the upmixer that can decode the peripheral channel information.

  A mixing factor can be created in a test downmixer environment, and the mixing factor depends on the input and output signal energy determined using the bandwidth limited (ie, filtered) input signal received at the test downmixer A test downmixer environment can be utilized to create The mixing factor determined using the test downmixer can then be used in the full bandwidth downmixer.

  A mixing coefficient value can be created by searching for a predetermined mixing coefficient value. The predetermined mixing coefficient value can be stored in the table format as a one-dimensional or two-dimensional table, for example, in the storage device of the downmixer. These tables can be indexed by the ratio of output energy to input energy. Retrieving one or more mixing coefficients from a mixing coefficient table to be used when downmixing the input signal when a substantially similar output-to-input ratio is encountered while downmixing the input signal Would be possible.

  Mixing coefficients can be created according to the input energy of multiple input channels. An energy ratio between at least one of the input channels and at least one other of the input channels can be determined, in which case mixing coefficient generation is responsive to the energy ratio. Mixing coefficient generation can include increasing one or more mixing coefficient values or decreasing one or more mixing coefficient values. Furthermore, the beginning of an acoustic event can be detected, in which case the mixing coefficient generation can be responsive to the input energy and the beginning of acoustic event detection.

  Other systems, methods, features and advantages of the present invention will be or will be apparent to those skilled in the art upon review of the following drawings and detailed description. Such additional systems, methods, features, and advantages are intended to be included herein, fall within the scope of the invention, and be protected by the following claims.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS A downmixer system is provided that generates mixing coefficients for downmixing a multichannel input signal having multiple input channels into an output signal having multiple output channels. The input energy level can be determined for at least a plurality of input channels, and at least one of the input energy level and the output energy level so that the signal energy and the intended direction of the input signal are substantially maintained. Depending on the determination, a blending coefficient can be generated. An output energy level can be determined for at least one of the output channels, depending on the input and output signal energy such that the signal energy and the intended direction of the input signal are maintained in the output signal. To generate a mixing coefficient.

  The number of output channels of the output signal may be smaller than the number of input channels of the input signal, for example, when a 3-channel input signal is downmixed into a 2-output channel output signal. For example, when a downmixer is used to downmix peripheral channel information, the number of input channels of the input signal may be equal to the number of output channels of the output signal.

  The downmixer can provide the listener listening to the output signal with a near accurate interpretation of the apparent direction and relative loudness of the input signal. When downmixing an input signal that contains both previous channel and peripheral channel information, the downmixer separately reduces the previous channel and peripheral channel information to maintain the input signal energy and desired direction in the output signal. Can be mixed. The downmixed peripheral channel information and the downmixed previous channel information can be combined (ie, combined) to create a two-channel mix of the input signal.

  When the downmixer downmixes the input multichannel signal to the output signal, the downmixer can change the energy ratio between the previous input channel and the peripheral input channel of the input signal. This energy ratio change can be used to provide the output signal with an approximately accurate interpretation of the reverberations present in the multichannel input signal. Energy ratio changes for downmixing can be achieved through mixing coefficient adjustment. In addition, the mixing factor can be adjusted to enhance acoustic events (i.e., sounds from instruments, syllables of words (phones), etc.). The acoustic event can occur in one or more input channels, such as the left and right peripheral channels, to provide a nearly accurate interpretation of the acoustic event in the output signal of the downmixer.

  A downmixer that downmixes an input signal having 3 input channels and an input signal having 5.1 input channels into an output signal having 2 output channels is discussed below. However, it is clear that the teachings described herein can be applied to input signals having different numbers of input channels and can be downmixed to output signals having more than two output channels. Will.

  FIG. 1 shows a function of a downmixing device capable of downmixing a multichannel input signal including at least three input channels into an output signal including a smaller number of output channels (here, two output channels) than the number of input channels. It is a block diagram. As shown in FIG. 1, the downmixer 100 includes a full bandwidth downmixer, generally indicated at 102, where the signal energy of the input signal and the desired direction are determined. In order to downmix the multi-channel input signal to the output signal in response to the generated left and right channel mixing factors ml and mr as substantially maintained in the output signal. The full bandwidth downmixer 102 can perform downmixing over a wide frequency range, such as a 20 to 20,000 frequency range. Other frequency ranges are possible. As described below, the downmixer 100 can further include a test downmixer 104 and a controller 106 that can be used to generate a test mix factor value. Can be used to update the left and right mixing factors ml and mr of the full bandwidth downmixer 102 so that the signal energy and intended direction of the input signal can be substantially maintained in the output signal. The test downmixer can operate over a limited frequency range, such as the 700-4000 Hz frequency range. Other frequency ranges are possible. The limited operating frequency range of the test downmixer may be advantageous as it allows the listener to create the full bandwidth downmixer 102 mixing factor using a particularly sensitive frequency range. Such a mixing coefficient generation method can take into account generation of a mixing coefficient that more accurately reflects in the output signal the loudness of the input signal perceived by the listener.

  Since the energy and intended direction are substantially maintained in the test downmixer 104 using the test mix factor, the input signal energy in the full bandwidth downmixer and if the test mix factor value is used in the full bandwidth downmixer It is possible to substantially maintain the desired direction in the output signal. Once the mixing factor is generated in the test downmixer 104 and the controller 106, the generated value can be used to update the mixing factor of the full bandwidth downmixer 102.

  As shown in FIG. 1, the full bandwidth downmixer 102 has an input with three channels to be downmixed (eg, left input channel (LI), center input channel (CI) and right input channel (RI)). The signal can be downmixed into an output signal having two channels (eg, a left output (LO) channel and a right output (RO) channel).

  The full bandwidth downmixer 102 includes a first mixer 108 and a second mixer 110, the first and second mixers for mixing the CI channel with the LI channel and the RI channel, the left channel mixing factor ml and Each of the mixing coefficients including the right channel mixing coefficient mr is designated. The CI channel can be mixed with the LI channel and the RI channel to create an L ′ channel and an R ′ channel, respectively. The first mixer 108 is coupled with a first phase shifter 112 for providing the desired phase shift to the L ′ channel to create the LO channel of the output signal. Similarly, the second mixer 110 is coupled with a second phase shifter 114 that adds the desired phase shift to the R ′ channel to create the RO channel of the output signal. Phase shifters 112 and 114 can provide pure phase shifts to the L ′ and R ′ channel information so that the energy and amplitude of L ′ and R ′ are not affected at any frequency.

The test downmixer 104 can include a first test mixer 116 and a second test mixer 118. The first test mixer 116 can receive at least one of the limited bandwidth (filtered) LI channel information LI _Lim and the CI channel information CI _Lim , and uses the test left channel mixing coefficient ml ′ and LI _Lim and CI _Lim channel information can be mixed to form a limited bandwidth test mixer left output channel LO _Lim . Similarly, the second test mixer 118, at least one can be received, _{RI Lim} and _{CI Lim} channel using the test right channel mixing coefficient mr 'of RI channel information _{RI Lim} and _{CI Lim} channel information limited bandwidth The information can be mixed to form the limited bandwidth RO output channel RO _Lim of the test mixer 104.

The controller 106 is coupled to the first mixer 108, the second mixer 110, the first test mixer 116 and the second test mixer 118. The controller 106 receives one or more of the LI, CI and RI channel information of the input signal and has a limited bandwidth (ie, filtered) such as LI _Lim , CI _Lim and RI _Lim used in the test downmixer 104. B) Channel information can be determined. The controller 106 may further receive output channel information such as output channel information LO and RO from the full bandwidth downmixer 102 and / or limited bandwidth output channel information LO _Lim and RO _Lim from the test mixer 104. The test downmixer 104 can be used to produce one or more mixing factor values, such as the mixing factor ml and mr of the full bandwidth downmixer 102, as described below. The controller 106 may further be coupled to a storage device 120, which provides one or more memory devices that may be utilized by the controller 106 as working memory and / or program memory, for example, during operation of the downmixer. To do.

  FIG. 2 shows a downmixer 100 when downmixing a multi-channel (ie> 2 channel) input signal (here having 3 channels) into an output signal having a smaller number of channels (here 2 channels) than the input signal. It is a flowchart which shows this operation | movement. As shown in FIG. 2, the input channel information is received 200 by the full bandwidth downmixer 102 as, for example, LI, CI, and RI channel information.

The controller 106 creates at least one of the mixing coefficients ml and mr used by the first mixer 108 and the second mixer 110 to mix the LI, CI and RI channel information using, for example, a test downmixer 104 as described below. 202. The full bandwidth downmixer 102 mixes 204 the LI and CI channels with the first mixer 108 and mixes the L ′ channel with L ′ = LI + ml ^* C (Equation 1)
Can be formed as

  The first phase shifter 112 can provide the desired phase shift to the L ′ channel information 206, and the resulting channel information is provided 212 as the LO channel of the output signal.

Similarly, the second mixer 110 mixes the RI and CI channels 208 and converts the R ′ channel to R ′ = RI + mr ^* C (Equation 2).
Can be formed as

  The second phase shifter 114 can provide 210 the desired phase shift to the R ′ channel information, and the resulting channel information is provided 212 as the RO channel of the output signal.

  Although generation 202 is shown as occurring at a particular location in the flowchart of FIG. 2, generation of mixing coefficients may be achieved at any time during operation of full bandwidth downmixer 102 and / or total bandwidth down. It will be apparent that this may be achieved at multiple intervals during operation of the mixer 102.

  The mixing factors ml and mr may be generated 202 at the same time or at different times during the operation of the full bandwidth downmixer 102. Furthermore, in some situations it may be desirable to generate only one mixing factor, ml or mr, utilized by the full bandwidth downmixer 102. Additionally or alternatively, during the mixing of the input signal, for example at regular time intervals (ie every 1.5 ms or every 10 ms), or a certain amount of input channel information (ie 64 samples or After processing 640 samples of input channel information), generation 202 can be achieved. Having generated one or both of the mixing coefficients ml and mr, the controller 106 can update the first and / or second mixers 108 and 110 with updated values for one or both of the updated mixing coefficients. . Such an update of the mixing coefficient value may occur at any time while downmixing the input signal to the output signal.

  The generation of the mixing coefficient will be generally described with reference to the flowchart of FIG. To illustrate the mixing coefficient generation for various conditions, the flowcharts and graphs of FIGS. 3-8 and FIGS. 11-13 are discussed in the context of FIG.

  FIG. 3 is a flowchart illustrating the generation 202 of mixing coefficients such as the left channel and right channel mixing coefficients ml and mr. The mixing coefficient generation may be performed by the test mixer 104 and the controller 106, for example. As shown in FIG. 3, at least one of the input channel energy and the output channel energy can be determined 300 by the controller 106 using, for example, a test downmixer 104. The controller 106 may determine 302 one or more feedback constants, for example, to smooth / stabilize mixing coefficient generation, particularly when there is rapidly changing input channel information. The controller can then generate 304 mixing coefficients such as test mixing coefficients ml ′ and mr ′ in response to the channel energy and / or feedback constant. The mixing factor of the total bandwidth downmixer 102 can be updated with the value of the test mixing factor.

  As described below, the controller 106 typically utilizes limited bandwidth input signal information by filtering the LI, CI and / or RI channel information to enhance audible frequencies in the frequency range of 700-4000 Hz, for example. To generate a mixing coefficient value. Filtering may emphasize other frequency ranges. Filtering the input channel can allow the generated mixing factor to more accurately reflect the loudness perceived by the listener. Although the full bandwidth downmixer 102 is typically a wideband downmixer that can downmix the input signal over a wide frequency range, eg, 20 Hz to 20 KHz, human hearing is an intermediate frequency, eg, a 700-4000 Hz frequency range. Is particularly sensitive to the energy content of the signal, and determining the mixing factor in response to the intermediate frequency range allows the listener to maintain the magnitude of the input signal at the most sensitive frequency. Is advantageous. Alternatively, or in addition, the controller 100 may generate mixing factor values using full bandwidth input channel information (ie, unfiltered input channel information).

  Below, the generation of one or more mixing factors in various situations is discussed. For example, FIGS. 4-6 are flowcharts illustrating the operation of the controller 106 that utilizes the test downmixer 104 to generate a mixing factor that can be used when downmixing a 3-channel input signal to a 2-channel output signal. FIG. 7 is a graph illustrating the mixing factor generated by the downmixer 100 according to the flowcharts of FIGS. 4-6 with a particular input signal such that the energy and intended direction of the input signal are substantially maintained in the output signal. It is. FIG. 8 is a graph showing the ideal mixing factor experimentally determined for a particular input signal such that the energy and intended direction of the input signal are substantially maintained in the output signal. The input signal scenario used when generating the graphs of FIGS. 7 and 8 can be used when generating a predetermined mixing coefficient value as will be described later. Other input signal scenarios may be used. FIGS. 11-13 show mixing coefficient generation for a downmixer that can downmix a 5.1 input channel to a 2 output channel.

  FIGS. 4-6 are flowcharts illustrating the generation of the mixing coefficients of FIG. 3 that can be utilized when downmixing a 3-channel input signal to a 2-channel output signal.

  FIG. 4 is a flowchart illustrating the operation of controller 106 and test downmixer 104 when determining 300 at least one of input and output channel energy. As shown in FIG. 4, input channel information including LI, CI and RI channel information is received 400 at the controller 106. Received input channel information 400 may include one or more digital signal samples of audio information received as an input signal representing at least one of LI, CI, and RI channel information.

The input channel information can be filtered 402 by the controller 106 to form limited bandwidth input channel information LI _Lim , CI _Lim and RI _Lim . For example, the input channel information may be filtered to emphasize substantially audible frequencies such as input signals in the frequency range from 700 to 4,000 Hz. The limited bandwidth input channel energy is then applied by the controller 106 for the LI and RI channels respectively ELI _Lim = LI _Lim ² + CI _Lim ² (Equation 3)
ERI _Lim = RI _Lim ² + CI _Lim ² (Equation 4)
Can be determined 404.

Limited bandwidth LO and RO channel information LO _Lim and RO _Lim in test downmixer 104 LO _Lim = LI _Lim + ml ′ ^* CI _Lim (Equation 5)
RO _Lim = RI _Lim + mr ' ^* CL _Lim (Equation 6)
Can be determined 406.

Limited bandwidth output channel energy by controller 106 for LO and RO channels respectively ELO _Lim = LO _Lim ² (Equation 7)
ERO _Lim = RO _Lim ² (Equation 8)
Can be determined 408. The limited bandwidth input and output channel energies determined at 404 and 408 are typically averaged by the controller 106 over multiple samples of input channel information received at the controller 106. The plurality of samples constitutes a first time period that can include, for example, 64 samples of the received 400 input channel information.

The limited bandwidth input and output channel energies are the total limited bandwidth energies for the LI _Lim , LO _Lim , RI _Lim and RO _Lim channels that can be averaged 410 as ELI _Sum , ELO _Sum , ERI _Sum , ERO _Sum channel energy, respectively. Where ELI _Sum = ELI _Sum + ELI _Lim (Equation 9)
ERI _Sum = ERI _Sum + ERI _Lim (Equation 10)
ELO _Sum = ELO _Sum + ELO _Lim (Equation 11)
ERO _Sum = ERO _Sum + ERO _Lim (Equation 12)
It is.

Next, it is determined 412 whether the averaging is complete. If it is determined 412 that averaging has not been completed, flow returns to receiving input channel information 400 as discussed above. However, if it is determined 412 that the first period has been completed, the total limited bandwidth input and output channel energy are respectively the total limited bandwidth left and right channel input and output energies EINL _Lim , EINR _Lim , EOUTL _Lim and EOUTR _Lim is determined 414 where EINL _Lim = ELI _Sum + ECI _Sum (Equation 13)
EINR _Lim = ERI _Sum + ECI _Sum (Equation 14)
EOUTL _Lim = ELO _Sum (Equation 15)
EOUTR _Lim = ERO _Sum (Equation 16)
It is.

  Once at least one of the input and output channel energies is determined at 300, a feedback constant can be determined 302 according to the flowchart of FIG.

FIG. 5 is a flowchart illustrating the operation of controller 106 when determining at least one feedback constant to generate a mixing factor that downmixes a three-channel input signal into two output channels. At 500, it is determined whether the total LO channel energy EOUTL _Lim is greater than the total limited bandwidth LI channel energy EINL _Lim . If it is determined 500 that the total limited bandwidth LO energy is not greater than the total limited bandwidth LI energy, the left channel feedback constant fbl is determined by the controller 106 as fbl = 0.98 ^* fbl (Equation 17)
Can be generated 502. The left channel feedback constant fbl can be initialized to a value of 1, for example. Other values, for example, between 0 and 1, may be used as the initial value of the feedback constant. However, if it is determined 500 that the total limited bandwidth LO channel energy is greater than the total limited bandwidth LI channel energy, the left channel feedback constant is determined by the controller 106 as fbl = 0.98fbl + gfb ((EOUTL _Lim / EINL _Lim ) -1) (Equation 18)
Is generated 504, where gfb can have the value 0.04. For example, if the value of gfb is large, the feedback loop may become unstable, and if the value of gfb is small, the feedback operation may be substantially reduced or eliminated. The value can be chosen experimentally.

Upon generating 502 or generating 504 a feedback constant, a determination 506 is made whether the total limited bandwidth RO channel energy EOUTR _Lim is greater than the total limited bandwidth RI channel energy EINR _Lim . If it is determined 506 that the total limited bandwidth RO channel energy is not greater than the total limited bandwidth RI channel energy, the right channel feedback constant fbr is determined by the controller 106 as fbr = 0.98 ^* fbr (Equation 19)
May be generated 510 as The value of fbr may initially be set as 1. However, if it is determined that the total limited bandwidth RO channel energy is greater than the total limited bandwidth RI channel energy, the right channel feedback constant fbr is determined by the controller 106 as fbr = 0.98 fbr + gfb ((EOUTR _Lim / EINR _Lim ) -1) (Equation 20)
May be generated 508.

  Although not shown, prior to determining one or both of the feedback constants fbl and fbr, the total limited bandwidth LO channel energy, the total limited bandwidth RO channel energy, the total limited bandwidth LI energy and / or the total It will be apparent that the limited bandwidth RI energy can be filtered (eg, low pass filtering). This filtering can be accomplished in the controller 106 as, for example, low pass filtering. For this low-pass filtering, for example, a time constant of 70 ms can be used. Other time constants may be used. Furthermore, it will be apparent that at least a portion of this filtering may not be performed by the controller 106 and that filtering may be accomplished by one or more filters embodied as a hardware device.

Returning to FIG. 3, when determining 302 the feedback constant, one or more test mixing factors may be generated 304 by the controller 106 as described with respect to the flowchart of FIG. 6. As shown in FIG. 6, 'ml by the controller 106' test left channel mixing coefficient ^{ml = 0.71 + fbl * lf +} fbr * rf ( equation 21)
, Where fbl and fbr have the values determined above with respect to FIG. 5, lf has a value of −1 and rf has a value of 0.3. The values of lf and rf can be used to bias the test mixing factors ml ′ and mr ′, respectively. When a 3 channel signal is downmixed and played through 2 channels, the test mix factor can be biased using, for example, lf and rf to compensate for localized (ie, desired direction) fine errors. it can. Other values of fl and rf may be used.

  After generating 600 the value of the test left channel mixing factor ml ′, the value of the test mixing factor ml ′ can be limited to a value between 0 and 1. For example, if ml ′ is determined to be less than 0, ml ′ is set to the value 0, and if ml ′ is determined to be greater than 1, ml ′ is set to the value 1.

Next, the test right channel mixing coefficient mr ′ is set by the controller 105 such that mr ′ =. 71 + fbl ^* rf + fbr ^* lf (Equation 22)
Can be generated 604, where fbl, fbr, rf and lf have the values discussed above with respect to generation 600.

  After the test mixing factor mr 'is generated, the value of mr' can be limited 606 to a value between 0 and 1. For example, if it is determined that the test blending coefficient mr ′ is less than 0, mr ′ can be set to a value of 0, and if the test blending coefficient mr ′ is determined to be greater than 1, the value of mr ′ is set. Can be set to 1.

  In order to substantially maintain the energy and intended direction of the limited bandwidth input signal received at the test downmixer 104 in the output signal of the test mixer, the test mixer downmixer left and right using, for example, the feedback constant fb Mixing coefficients ml ′ and mr ′ have been determined. Since the energy and intended direction is substantially maintained in the test downmixer 104 using the test mix factor, the test mix factor value, if used in the full bandwidth downmixer 102, is Allows the energy and intended direction of the input to be substantially maintained in the output signal. The test mix coefficient values ml 'and mr' can be used to update 608 the mix coefficient values ml and mr used in the full bandwidth downmixer 102.

  The update 608 may be accomplished by the controller 106 updating the left channel mixing factor ml of the first mixer 102 with the value of the test left channel mixing factor ml 'by replacing the value of ml with the value of ml'. Similarly, the right channel mixing coefficient mr is updated by the controller 106 which updates the right channel mixing coefficient mr of the second mixer 104 with the value of the test right channel mixing coefficient mr ′ by replacing the value of mr with the value of mr ′. Can do.

  In addition, or alternatively, the left and right channel mixing coefficients are updated 608 by the controller 106 by smoothing the mixing coefficients before they are used in the full bandwidth downmixer that actually outputs the signal. can do. This smoothing may be performed at a time between the calculation of new values of ml and mr. For example, the value of ml in the full bandwidth downmixer can be changed (ie, updated) approximately every 0.5 seconds to approach the calculated value ml '. This change is made so that the ml reaches the value of ml 'in the full bandwidth downmixer before the other value of ml' is determined by the test downmixer 104. The same may be true for the mixing coefficient value mr being updated with the test mixing coefficient value mr '.

  In this way, left and right channel mixing factors ml and mr can be generated 304 for the full bandwidth downmixer 102.

  FIG. 7 is a graph of mixing coefficients that can be generated by the downmixer 100 according to the flowcharts of FIGS. 4-6 for a single input signal applied to the CI and LI channels. The graph of FIG. 7 is generated by a single signal that is smoothly panned between the LI and CI channels, in which case the intended direction of the input signal is known precisely. FIG. 8 is a graph of the mixing factor as a function of the panning angle obtained experimentally to compensate for the fine error of localization when a 3 channel signal is downmixed and reproduced through 2 channels. The graph in FIG. 8 shows the calculated ideal case, where there is a single signal that is smoothly panned between the LI and CI channels, and the intended direction of the input signal is precisely I know it. The left channel mixing coefficient ml value is shown using a broken line in FIGS. 9 and 10, and the right channel mixing coefficient mr value is shown using a solid line in FIGS.

  For example, mixing coefficients such as ml and mr may be generated 202 as a predetermined value depending on the input channel energy (FIG. 2) and need not be generated in real time. Such a scheme can utilize frequency limited input and output energy from the test downmixer as input to one or more one or two dimensional lookup tables. As is apparent from the previous description of the downmixer operation, the mixing factor may depend on the ratio of input energy to output energy. Mixing coefficients such as ml and mr can be directly derived using a look-up table with the output / input energy ratio determined by the test downmixer as input.

  The controller 106 and a downmixer, such as the downmixer 102 or the test downmixer 104, can be used to generate the predetermined mixing factor stored in such a lookup table, such as the mixing factor ml and mr. In this case, the input signal for a particular input signal scenario (ie, having a smooth panning characteristic from CI to LI, for example, used to generate the graph of FIG. 8) is processed by a downmixer. The ratio of the output energy and the input energy derived from the input signal scenario can be determined. Thereafter, at least one of, for example, mixing coefficients ml and mr, which can be used in the specific input signal scenario such that the signal energy in the intended direction of the input signal is substantially maintained in the output (downmixed) signal. The downmixer and controller 106 can be utilized to determine the mixing factor. For example, the mixing factor can be generated as discussed above with respect to FIGS.

  The ratio of output and input energy for a particular input signal scenario can be stored in the storage device 120 in tabular format. Such a table format may include, for example, mixing coefficients ml and mr indexed by the ratio of output energy to input energy for one or more input signal scenarios. For example, a mixing coefficient table for ml can be provided for a particular input signal scenario and indexed by the ratio of output energy to input energy. Similarly, a mixing coefficient table for mr can be provided for that particular input signal scenario and indexed by the ratio of output and input signal energy.

  In operation, the controller 106 is used by the downmixer to detect a particular input signal scenario, determine a ratio of output and input energy, and based on that ratio, downmix the signal for that input signal scenario. A reference value can be determined for at least one mixing factor such as, for example, the mixing factors ml and mr. The blending factor that is retrieved allows the input energy and intended direction of the input signal to be substantially maintained in the output signal. The controller may update the downmixer mixing coefficient value with the retrieved mixing coefficient value, for example, in a manner similar to that discussed above with respect to update 608 of FIG.

  In this way, a library of predetermined mixing coefficient scenarios can be determined and stored in the storage device 120, for example. The library can include a blending coefficient table for blending coefficients, each blending coefficient table providing one or more blending coefficients that are indexed by a ratio of output energy to input energy. Other mixing coefficient table configurations are possible. The blending coefficient library may be accessed by a controller that retrieves blending coefficient values for a particular input signal scenario.

  Predetermined mixing factor generation may be utilized in conjunction with the mixing factor generation described above with respect to FIGS. For example, the controller may attempt to check whether the input signal meets the requirements for a particular input signal scenario such that the mixing coefficient library includes a predetermined mixing coefficient for it. If the controller 106 determines that the input signal matches one of the input signal scenarios for which the mixing factor is stored, the controller retrieves the appropriate mixing factor from the mixing factor library as described above. Thus, the mixing coefficient can be generated. However, if it determines that the input signal does not meet the criteria for the stored input signal scenario, the controller, in conjunction with the test mixer 104, generates a mixing factor for the downmixer.

  In addition or alternatively, a learning algorithm that allows the controller to identify features for an input signal scenario where a predetermined mixing factor is useful (ie, an input signal scenario that is repeatedly received with an input signal in a downmixer) The controller can be used. In such a case, the controller can use a test downmixer to determine the mixing factor value stored in the storage device 120 for that particular input signal scenario. When subsequently recognizing the input signal scenario, the controller 106 can generate a mixing factor for that scenario by retrieving the mixing factor from the mixing factor table.

  By generating the mixing factor value by retrieving the mixing factor as described above, the controller can reduce the downmixer resource requirements than would be required to generate the mixing factor as described above with respect to FIGS. A quantity can be generated that allows the input signal energy and the desired direction to be maintained in the output signal. The downmixer resource may be freed so that it can be used in other operations of the downmixer.

  FIG. 9 is a block diagram of a downmixer 900 according to the present invention. The downmixer 900 can receive a multi-channel input signal including three or more channels and downmix the multi-channel input signal into an output signal including a smaller number of channels than that of the input signal. The downmixer 900 includes at least one of the front channel left and right mixing factors ml and mr and the surrounding channel mixing factors mi and ms so that the energy and intended direction of the input signal are substantially maintained in the output signal. It includes a full bandwidth downmixer 901 for downmixing a 5.1 channel input signal into a 2 channel output signal using one. The downmixer 900 further includes a test downmixer 104 'that can be utilized in connection with the controller 940 when generating the front channel left and right mixing coefficients ml and mr. The operation of the test mixer 104 'will not be discussed in detail because the pre-channel mixing coefficients ml and mr can be generated in the same way that the mixing coefficients ml and mr are generated by the test mixer 104 and controller 106 of FIG. The downmixer 900 is one of the peripheral mixing factors such as the peripheral mixing factors mi and ms so that the signal energy and the intended direction of the input signal are substantially maintained in the output signal of the full bandwidth downmixer 901. It further includes a test downmixer 950 that can be utilized with the controller 940 when generating one or more.

  As shown in FIG. 9, front left input (LI), front center input (CI), front right input (RI), low frequency (LFE), left peripheral input (LSI) and right peripheral input (RSI) A channel may be received at the downmixer 900. The downmixer 900 can downmix the 5.1 input channel of the input signal into an output signal including, for example, two output channels, a left output (LO) and a right output (RO) channel.

  The full bandwidth downmixer 901 can include a first LI mixer 902 for mixing the LI, CI and LFE channels and a first RI mixer 904 for mixing the RI, CI and LFE input channels of the input signal. . Multipliers 906 and 908 may be used to multiply the CI input signal by the front left and right channel mixing coefficients ml and mr, respectively, before mixing the CI channels in the first LI mixer 902 and the first RI mixer 904. The second LI mixer 9010 can allow one or both components of the peripheral channel LSI and RSI to be added to the LI ′ channel information to achieve the desired phase shift to form the LO ′ channel information. LI phase shifter 912 may be provided. Similarly, a second RI mixer 914 can be provided to add one or both peripheral channel LSI and RSI components to the RI ′ channel information to achieve the desired phase shift to form the RO ′ channel information. An RI phase shifter 916 can be provided.

  An LSI mixer 918 can be provided to add the components of the RSI channel to the LSI channel, and a multiplier 922 to provide an LSI mixing factor (eg, a mi peripheral mixing factor corresponding to the imaginary component SLI ′ of the LO channel). Can be provided. An LSI phase shifter 924 can be provided to provide a desired phase shift to the LSI 'channel information to form the LSO' channel information. Similarly, an RSI mixer 930 can be provided to add the LSI channel components to the RSI channel, allowing the multiplier 932 to supply the mi peripheral mixing factor and form the RSO ′ channel information to form the RSO ′ channel information. An RSI phase shifter 934 can be utilized to provide the desired phase shift to the channel information.

  Multipliers 919 and 921 can be provided to provide the ms peripheral mixing factor. For example, the ms marginal mixing factor can be used to control the amount of LSI and RSI channels applied to each previous channel output path (eg, to the LI 'and LO' signals, respectively).

  An LO mixer 936 may be provided to mix the LSO 'and LO' channel information to form an output signal output channel LO. Similarly, RO mixer 938 can be utilized to mix RO 'and RSO' channel information to form the RO output channel of the output signal.

  The test downmixer 950 can include a first test adder 952 and a second test adder 954. The first test adder 952 is coupled to the first test mixer 956 and the second test mixer 958 to provide test peripheral mixing coefficients mi ′ and ms ′ in the test mixer 950. Similarly, the second test adder 954 is further coupled to a third test mixer 960 and a fourth test mixer 962 that can supply test peripheral mixing coefficients ms 'and mi', respectively, in a test down mixer 950.

  The controller 940 utilizes, for example, LSI, LI, CI, LFE to generate and / or update one or more of the mixing coefficients ml, mr, ms, and mi utilizing the test downmixers 140 ′ and 950. , RI, and RSI input channels, and one or more of the multipliers 906, 908, 919, 921, 922, and 932 of the full bandwidth downmixer 901. In order to reduce confusion, the coupling between the controller 940 and the multipliers 906, 908, 919, 921, 922, and 932 is shown with dashed lines.

The first test adder 952 can receive limited bandwidth (ie, filtered) LSI channel information as an LSI _Lim received at the test downmixer 950 and attenuated by a factor of 0.91. The first test adder 952 can further receive the RSI limited bandwidth channel information as an RSL _Lim that has been inverted and multiplied by the cross-correlation factor −0.38, and can add this to the attenuated LSI _Lim signal. . The resulting channel information from the first test adder 952 is mixed in the first and second test mixers 956 and 958 according to the test peripheral mixing factors mi ′ and ms ′ to be mixed into the test mixer 950 output channel information LSO-Im. _Lim and LSO-Re _Lim can be generated, respectively. Similarly, the second test adder 954 can add the inverted RSI _Lim channel information attenuated by a factor of 0.91 with the LSI _Lim channel information multiplied by the cross-correlation factor -0.38. . The resulting channel information is mixed in the third and fourth test mixers 960 and 962 in accordance with the test peripheral mixing coefficients ms ′ and mi ′ to generate test mixer 950 output channel information RSO-Re _Lim and RSO-Im _Lim , respectively. can do.

The controller 940 can be further coupled to the test down mixer 104 ′, the first, second, third and fourth test mixers 956, 958, 960 and 962. The controller 940 can receive one or more of the LI, CI, RI, LFE, LSI, and RSI channel information of the input signal, and also includes, for example, LSI _Lim and RSI _Lim used in the test downmixer 950. Limited bandwidth (ie, filtered) channel information can be determined. The controller 940 further includes, for example, output channel information LO and RO from the full bandwidth downmixer 901 and / or limited bandwidth output channel information from the test downmixer 950 LSO-IM _Lim , LSO-RE _Lim , RSI-RE _Lim And output channel information such as RSI-IM _Lim channel information and using a test downmixer 950 as described below, for example, one or more mixing factors such as mixing factors ml, mr, mi and ms Can be generated. Further, the controller 940 can be coupled to a storage device 942 that provides working memory and program memory for the controller 940. The operation of the downmixer 900 will be discussed with reference to the flowchart of FIG.

  FIG. 10 is a flowchart showing the operation of the downmixer 900 of FIG. As shown in FIG. 10, for example, input channel information including information for LSI, LI, CI, LFE, RI, and RSI channels of an input signal is received 1000. Creating 1002 one or more mixing coefficients using controller 940 and test downmixer 950 in response to at least one of the input channel information as described below with reference to FIGS. 11-13 and 14-17. Can do. As discussed above with reference to FIGS. 3 and 4-6, LI, CI, LFE and RI channel information can be similarly mixed 1004. Further, the LFE channel information can be amplified, for example, by a factor of 2, before being mixed by the first LI and RI mixers 902 and 904, respectively. In addition, before the CI channel information is mixed in the first LI and RI mixers 902 and 904, using the multipliers 906 and 908, the CI channel information is one or more of the front left and right channel mixing factors ml and mr, for example. A mixing factor can be supplied. The first LI mixer 902 generates LI 'channel information, and the first RI mixer 904 generates RI' channel information. For example, in a manner similar to that discussed above with respect to FIGS. 3-11, the LI 'and RI' channel information can be utilized as the left and right output signals for the purpose of producing the mixing coefficients ml and mr.

  The second LI mixer 910 and the second RI mixer 914 can be used to add 1006 the LSI and RSI channel components to the LI 'and RI' channel information, respectively. For example, the LSI channel information can be multiplied by the mixing coefficient ms by the multiplier 919 before being mixed with the LI ′ channel information by the second LI mixer 910. Similarly, the RSI channel information can be multiplied by the mixing factor ms at the multiplier 919 before being mixed with the RI ′ channel information at the second RI mixer 914. A desired phase shift for the previous channel information may be provided 1008 by LI phase shifter 912 and RI phase shifter 916 to form LO 'and RO' channel information, respectively.

  The RSI and LSI channel components can be added 1010 at the same time as, or after, mixing 1004, addition 1006 and provision 1008. For example, the RSI channel can be inverted by an inverter 927, multiplied by a cross-correlation factor of −0.38, for example, by a multiplier 928, and mixed with LSI channel information by an LSI mixer 918. Prior to mixing by the LSI mixer 918, the LSI channel information can be attenuated by a multiplier 929 by a factor such as 0.91. Similarly, the LSI channel information can be multiplied by, for example, a cross-correlation factor of −0.38 using the multiplier 931 and mixed with the RSI signal by the RSI mixer 930 to add the LSI channel component to the RSI channel. Prior to mixing with the RSI mixer, the RSI channel can be attenuated with a multiplier 933 by a factor of, for example, 0.91.

  By multiplying the channel information from the LSI mixer 918 and the RSI mixer 930 by the mixing coefficient mi 1012, the respective mixing coefficients can be supplied to form LSI ′ and RSI ′ channel information, respectively.

  Any desired phase shift can be provided 1014 for the peripheral channels. For example, the LSI phase shifter 924 can phase shift the LSI ′ channel information to form LSO ′ channel information, where the phase is offset by 90 degrees relative to the phase provided by the LI phase shifter 912. . Similarly, the phase of RSI ′ channel information can be changed by RSI phase shifter 934 to form RSO ′ channel information, where the phase shift is 90 degrees offset from that used by RI phase shifter 916. Is done.

  Thereafter, the peripheral channel information and the previous channel information are mixed 1016. For example, the LO mixer 936 can mix the LSO 'channel information with the LO' channel information to form the LO channel of the output signal and provide the LO channel 1018. Similarly, RSO 'channel information can be mixed with RO' channel information at RO mixer 938 to form the RO channel of the output signal, and LO channel can be provided 1018.

  Although the step of creating the mixing factor is shown at a particular location in the flowchart of FIG. 10, one or more mixing factors such as ml, mr, mi, and ms may be applied at any point during the operation of the downmixer 900. It will be apparent that the controller 940 can make it. Further, it is not always necessary to create all the mixing coefficients at the same time, and they can be created at various times during the operation of the downmixer 900. The front left and right channel mixing factors ml and mr can be made using the controller 940 and test downmixer 104 ′ in a manner similar to that discussed above with respect to FIGS. 3 and 4-6, and will not be discussed in detail. . Furthermore, the mixing coefficient generation for the previous channel mixing coefficients ml and mr can be achieved separately from the mixing coefficient generation of the peripheral mixing coefficients mi and ms.

  For example, as discussed with respect to the flow chart of FIG. 3 and the flow charts of FIGS. 11-13 and 14-17, the controller 940 can use the test mixer 950 to generate the generation of the ambient mixing factors mi and ms. As shown in FIG. 3, at least one of the input and output channel energies can be determined 300. The determination 300 of at least one of input and output channel energy is discussed with respect to the flowchart of FIG.

  FIG. 11 is a flowchart illustrating the operation of the controller 940 when determining the input channel energy used to generate at least one test peripheral mixing factor such as, for example, the test peripheral mixing factors mi ′ and ms ′. As shown in FIG. 11, input channel information for LSI and RSI channels is received at controller 940, for example, as signal samples of the input signal in a manner similar to that discussed above with respect to receiving 400 of FIG. 1100.

The input channel information can be filtered 1102 by the controller 940 to generate limited bandwidth input channel information LSI _Lim and RSI _Lim channel information. In a manner similar to that discussed above with respect to filtering 402 of FIG. 4, the input channel information can be filtered 1102 utilizing, for example, a finite impulse response filter that emphasizes the frequency and 700-4000 Hz frequency range.

The limited bandwidth output channel information is converted into LSO real and imaginary channel information LSO-Re _Lim and LSI-Im _Lim and RSO real and imaginary channel information RSO-Re _Lim and RSO-Im _Lim in test downmixer 950 LSO-Re _Lim = ms ' ^* LSI _Lim (Equation 23)
LSO-Im _Lim = mi ′ ^* (0.91 ^* LSI _Lim + .38 ^* RSI _Lim )
(Equation 24)
RSO-Re _Lim = ms' ^* RSI _Lim (Equation 25)
RSO-Im _Lim = mi ' ^* (-.91 ^* RSI _Lim- .38 ^* LSI _Lim )
(Equation 26)
Can be determined 1104, where ms 'and mi' are initialized to a value of 0.7. Limited bandwidth input channel energy can be determined 1106 for LSI energy and RSI energy by controller 940 as ELSI _Lim and ERSI _Lim, respectively, where ELSI _Lim = ELSI ² _Lim (Equation 27)
ERSI _Lim = ERSI ² _Lim (Equation 28)
It is.

The controller 940 causes the limited bandwidth output channel energy to be LSO channel energy real and imaginary components ELSO-Re _Lim and ELSO-Im _Lim , and RSO channel energy real and imaginary ERSO-Re _Lim and ERSO-Im _Lim. Decision 1108 can be made, where ELSO-Re _Lim = LSO-Re ² _Lim (Equation 29)
ELSO-Im _Lim = LSO-Im ² _Lim (Equation 30)
ERSO-Re _Lim = RSO-Re ² _Lim (Equation 31)
ERSO-Im _Lim = RSO-Im ² _Lim (Equation 32)
It is.

For example, in a manner similar to that discussed above with respect to averaging 410, the controller 940 reduces the limited bandwidth input and output channel energy as LSI, RSI, LSO and RSO average energies ELSI _Sum , ERSI _Sum , ELSO _Sum , and ERSO _Sum. Can be averaged 1110, where ELSI _Sum = ELSI _Sum + ELSI _Lim (Equation 33)
ERSI _Sum = ERSI _Sum + ERSI _Lim (Equation 34)
_{ELSO Sum = ELSO Sum + ELSO-} Re LIm + ELSO-Im Lim
(Equation 35)
ERSO _Sum = ERSO _Sum + ERSO-Re _Lim + ERSO-Im _Lim
(Equation 36)
It is.

A determination 1112 can be made whether the averaging is complete. If averaging has not been completed, flow returns to receipt 1100. If it is determined that averaging has been completed 1112, the controller sets the total limited bandwidth input and output channels as EIn _Lim and EOut _Lim , respectively, EIn _Lim = ELSI _Sum + ERSI _Sum (Equation 37)
EOut _Lim = ELSO _Sum + ERSO _Sum (Equation 38)
Can be determined 1114.

  Returning to FIG. 3, when determining 300 at least one of the input and output channel energies, a feedback constant can be determined 302. Feedback constant determination 302 is discussed with respect to the flowchart of FIG.

FIG. 12 is a flowchart illustrating the operation of the controller 940 when determining a feedback constant fbsi that may be used when determining a test mixing factor for the test downmixer 950, such as, for example, test peripheral channel mixing factors mi ′ and ms ′. It is. As shown in FIG. 12, for example, the limited bandwidth input and output energy determined at 1114 is filtered 1200 by the controller 940 to filter the input and output limited bandwidth energy SIN _Lim and SOUT _Lim SIN _Lim = . 98 ^* SIN _Lim +. 02 ^* EIN _Lim (Equation 39)
SOUT _Lim =. 98 ^* SOUT _Lim +. 02 ^* EOUT _Lim
(Equation 40)
Can be formed as This filtering may be low pass filtering and may be accomplished using a filter having a time constant of 70 ms, for example. Other time constants may be used.

The feedback constant fbsi is changed by the controller 940 to fbsi =. 98 ^* fbsi + gfb ^* ((SOUT _Lim / SIN _Lim ) -1)
(Equation 41)
Can be determined 1202 where gfb has a value of 0.04. Considerations for the value of gfb to be used may be similar to those discussed above with respect to generation 504 discussed above with respect to FIG. Once the feedback constant is determined 302, one or more test marginal mixing factors can be generated 304 by the controller 940, as described with respect to FIG.

FIG. 13 is a flowchart illustrating the operation of the controller 940 when generating test peripheral mixing factors for the downmixer 900, such as, for example, test peripheral channel mixing factors mi ′ and ms ′. As shown in FIG. 13, it is determined 1300 whether the value of the feedback constant fbsi determined at 1202 is greater than or equal to zero. If the feedback constant is neither greater than zero nor equal to zero, the value of the test ambient mixing factor ms ′ is 1302 by the controller 940 at ms ′ = 0−fbsi (Equation 42)
And the value of the test mixing factor mi ′ is set to 1 at 1304. However, if it is determined 1300 that the feedback constant is greater than or equal to zero, the value of ms ′ is set to zero at 1306 and the value of mi ′ is set to mi ′ = 1−fbsi at 1308 (Equation 43 )
Set to If mi ′ is less than zero, mi ′ is reset to a zero value at 1310.

  After generating the test mixing factors mi ′ and ms ′, the test peripheral mixing factors mi ′ and ms ′ may be utilized by the controller to update the peripheral mixing factors mi and ms used by the full bandwidth downmixer 901. it can. Update 1312 may be accomplished in a manner similar to that described above, for example with respect to update 608 of FIG.

  For example, when LSI or RSI channels are driven together by the same signal, the mixing factor mi can be utilized in the downmixer 900 to attenuate one or both of the peripheral channels. The peripheral mixing factor mi can be adjusted by a small feedback loop to keep the input power and output power substantially equal. A marginal mixing factor ms can be used to bypass the 90 degree phase shifters 924 and 934, where ms is the cross-mix applied to the previous channel, for example when the LSI and RSI are out of phase. The amount of peripheral signal generated can be controlled. When ms has a positive, non-zero value, the coherent signal of the peripheral input channel can be provided in both the 90 degree phase shift path and the non-90 degree phase shift path of the downmixer 900.

  At least in some circumstances, for one or more peripheral mixing factors such as, for example, the peripheral mixing factors mi and ms determined with respect to FIG. 13, they are changed before or during use by the downmixer 900. It may be desirable to make adjustments. Similar to the generation of the pre-channel mixing factors ml and mr, the peripheral channel mixing factors mi and ms are usually generated in a test downmixer environment. By utilizing a test downmixer to generate one or both of the mixing coefficients mi and ms, the coefficient can be further modified / adjusted before it is used in the full frequency range downmixer, If so, the values of mi and ms can be maintained in the test downmixer so as not to interfere with the feedback.

  By actively changing the energy ratio between the front channel and the rear channel, to create a two-channel downmix that is subjectively closer to the original five-channel downmix, one or both of the peripheral mixing factors mi and ms The value can be adjusted. Such changes can be adjusted to suit when there is an excessive response in the peripheral channels. The mixing coefficients mi and ms can be adjusted using the ratio F / S of the energy of the front channel and the peripheral channel. As will be discussed below, this adjustment is achieved by reducing at least one or both of mi and ms by a certain amount (eg, an amount corresponding to 3 dB of LSI and / or RSI channel information) when the F / S ratio is greater than 1. It may include reducing. Furthermore, in some situations it may be desirable to actively look for audible acoustic elements (ie, non-resonant acoustic information) residing on one or more of the input channels (eg, one or both of the peripheral channel LSI and RSI). Let's go. When an audible acoustic element is present, the 3 dB attenuation added to the mixing factors mi and ms can be removed.

  In addition, various acoustic events may be adjusted by adjusting the peripheral mixing factors mi and ms, for example, by emphasizing peripheral channel signals that may not be as strong as the simultaneous signals generated in the previous channel received by the downmixer 900. Can be strengthened. An acoustic event may be thought of as a directional transient, such as a sound that has an initial energy spike and retains information about the direction of the transition (ie, unobstructed by an object), such as screaming or drumming. . The two types of acoustic events will be syllables and impact sounds. Syllables can include phonemes and sounds. Phonemes are transient sounds that can be particularly useful in detecting and localizing syllables in a person's story as well as showing the characteristics of a single phone in the person's story. A sound is an individual sound made by a musical instrument. Since sounds and phonemes have common characteristics, they can be collectively referred to as “syllables”. Syllables generally have the following characteristics: have a finite duration ranging from approximately at least 50 ms to about 200 ms, usually on the order of 150 ms; have a rise time of about 33 ms; generally from about 0.2 ms to about 0.5 ms It does not occur more frequently than the frequency of times; it can have less or greater volume (amplitude). In contrast, an impact sound would be a transient sound with a very short duration, such as drum or flicks or explosive sounds in the talk. Impact sounds generally have the following characteristics: a short duration from about 5 ms to about 50 ms, a rise time from about 1 ms to about 10 ms, and a loud volume.

  For example, David H., filed on May 2, 2003 as agent docket number 11336/208, commonly assigned and incorporated herein by reference. Acoustic events can be detected as described in US Patent Application No. (not yet assigned) entitled “Acoustic Event Detection” by Glee Singer. For example, the rate of increase of the input energy level in one of the input channels can be used to detect the beginning of an acoustic event. For example, the rate of increase in one or both of the LSI and RSI channels can be detected, in which case the mixing factor mi and so that the acoustic event can be more noticeable in two-channel mixing than if the signal power is fully maintained. The value of ms can be adjusted. For example, the 3 dB attenuation used to remove the reverberations detected in one or more of the input channels can be removed. An acoustic detector may be utilized in connection with any input channel, and the presence of a significant acoustic event in a particular input channel can be used to trigger a temporary boost in that channel's level. The boost may be achieved by increasing the value of one or more mixing factors such as mixing factors mi and ms. Such a boost may last for example from 100 to 300 ms. Further, to increase the audibility of low level acoustic events in the resulting downmix, the boost may be, for example, a boost corresponding to 1-3 dB of the corresponding channel information.

  14 to 17 are flowcharts showing adjustment of the peripheral mixing coefficient.

  FIG. 14 is a flowchart illustrating the operation of the controller 940 when adjusting one or more mixing coefficients such as, for example, the peripheral mixing coefficients mi and ms. As shown in FIG. 14, the input channel energy is determined 1400. Input channel energy determination 1400 will be discussed later with respect to the flowchart of FIG. Once the input channel energy is determined 1400, one or more mixing factors such as mi and ms can be adjusted 1402, for example. Mixing factor adjustment 1402 will be discussed later in connection with the flowcharts of FIGS.

FIG. 15 is a flowchart illustrating the operation of the controller 940 when determining 1400 the input channel energy. Input channel information is received 1500 and may include information regarding the LI, RI, CI, LSI and RSI channels of the input signal. The previous input channel energy can be determined 1502 as ELI, ECI and ERI for LI, CI and RI channels, where ELI = LI ² (Equation 44)
ECI = CI ² (Equation 45)
ERI = RI ² (Equation 46)
It is.

IP channel information may be received 1500 in a manner similar to that described above with respect to receiving 400 of FIG. The total previous input channel energy can be determined 1504 as EFI, where EFI = ELI + ECI + ERI (Equation 47)
It is.

Peripheral input channel energy can be determined 1506 for ELSI and RSI channels as ELSI and ERSI, respectively, where ELSI = LSI ² (Equation 48)
ERSI = RSI ² (Equation 49)
It is.

Total peripheral input channel energy ESI ESI = ELSI + ERSI (Equation 50)
As 1508 can be determined.

The previous and peripheral input channel energies can be averaged 1510 as EFI _Sum and ESI _Sum , respectively, where EFI _Sum = 0.9 ^* EFI _Sum + 0.1 ^* EFI (Equation 51)
^{_{ESI Sum = 0.9 * ESI Sum +0.1}} * ESI ( equation 52)
It is.

  Averaging 1510 may be accomplished in a manner similar to that discussed above, for example with respect to averaging 410 in FIG.

A determination 1512 can be made whether the averaging is complete. If averaging has not been completed, flow returns to receiving input channel information 1500 and continues as discussed above. If it is determined 1512 that the averaging is complete, the previous and peripheral input channel averages are filtered 1514 as EFI _Lim and ESI _Lim , where EFI _Lim = 0.99 ^* EFI _Lim + 0.01 ^* (EFI _Sum )
(Equation 53)
ESI _Lim = 0.97 ^* ESI _Lim +0.03 ^* (ESI _Sum )
(Equation 54)
It is.

  Once the input channel energy is determined 1400, the mixing factor can be adjusted 1402 as described above with respect to the flowcharts of FIGS.

FIG. 16 is a flowchart illustrating the operation of the controller 940 when adjusting 1402 one or more mixing coefficients such as, for example, the peripheral mixing coefficients mi and ms. As shown in FIG. 16, the ambient energy boost factor SBF is expressed as SBF = 3 ^* ESI-2 ^* ESI _Lim (Equation 55)
Can be generated 1600 as

Next, it can be determined whether the average peripheral energy ESI _Lim is increasing. This is accomplished by determining 1602 whether the average acoustic energy is less than the ambient energy boost factor. When it is determined that the average ambient energy is less than the ambient energy boost factor, for example, ESI _Sum = 0.99 ^* ESI _Sum +0.01 SBF (Equation 56) using the first time constant
The average ambient energy can be averaged 1604. The first time constant may be about 150 ms, for example.

However, if it is determined 1602 that the average ambient energy is not less than the energy boost factor, then the average ambient energy is calculated using the second time constant: ESI _Sum = 0.999 ^* ESI _Sum +0.001 SBF (Equation 57 )
Can be averaged 1606. Here, the second time constant may be about 1.5 seconds, for example.

  The average peripheral input energy may be averaged according to the current value of the peripheral input energy. This may be accomplished, for example, by steps 1602, 1604, and 1606.

The front / rear energy ratio F / S is the energy ratio of the average front channel and average peripheral channel input energy: F / S = (EFI _Sum +1) / ((1.2 ^* ESI _Sum ) +1)
(Equation 58)
It may be determined 1608. This pre / post energy ratio may be a bias to the peripheral input channel, for example 1.2 dB. Further, the pre / post energy ratio may be limited to within the range of 0.1 and 10. For example, if the front / rear power ratio is greater than 10, the front / peripheral energy ratio may be set to a value of 10. If the front / peripheral energy ratio is less than 0.1, the front / peripheral energy ratio may be set to a value of 0.1.

The mixing factors mi and ms may be determined according to the previous / ambient energy ratio. This can be accomplished by determining 1610 whether the front / ambient energy ratio is greater than four. If the front / ambient energy ratio is greater than 4, the mixing coefficients ms and mi are 1612 and 1614 ms = 0.71 ^* ms (Equation 59)
mi = 0.71 mi (Equation 60)
It can be set to.

However, if it is determined 1610 that the front / peripheral energy ratio is not greater than 4, it may be determined 1616 whether the front / peripheral energy ratio is 2 or more and 4 or less. If the front / peripheral energy ratio is 2 or more and 4 or less, the mixing coefficients ms and mi are 1618 and 1620, respectively. Ms = 0.8−0.045 ^* (F / S−2) (Equation 61)
mi = 0.8−0.045 ^* (F / S−2) (Equation 62)
Set 1618 and 1620 may be used.

However, if it is determined 1616 that the front / peripheral energy ratio is not less than 2 and not less than 4, the mixing coefficients ms and mi are ms = 1−0.2 ^* (F / S−1) (Equation 63)
mi = 1-0.2 ^* (F / S-1) (Equation 64)
Set 1622 and 1624 may be used.

  Further, for example, the values of the mixing coefficients such as the peripheral mixing coefficients mi and ms may be adjusted as the peripheral channel level increase ratio S / I according to the increase of the peripheral channel input level. Adjustment to the mixing factors mi and ms as a function of the rear peripheral channel input level will be discussed with respect to the flowchart of FIG.

FIG. 17 is a flowchart illustrating the operation of the controller 940 when adjusting one or more mixing coefficients such as the peripheral mixing coefficients mi and ms according to the rear peripheral input energy level ratio S / I. As shown in FIG. 17, a post-peripheral input energy ratio S / I is generated 1700, where S / I = SBF / ESI _Lim (Equation 65)
Where the ambient energy boost factor is as determined with respect to FIG. 16 and the ESI _Lim is as determined with respect to FIG. Next, a determination 1702 is made whether the second peripheral boost factor indicator SBF2 is less than the peripheral input energy ratio. If the second boost factor is less than the energy ratio, the second peripheral boost factor is SBF2 = 0.8SBF2 + 0.2S / I (Equation 66)
As a set 1704.

However, if the second peripheral boost factor is not less than the peripheral input energy ratio, the second peripheral boost factor indicator is SBF2 = 0.97SBF2 + 0.03S / I (Equation 67)
1706, where the second peripheral boost factor 1704 represents a time constant of about 7 ms, and the second boost factor at 1706 represents a time constant of about 70 ms.

The second peripheral boost factor indicator may be scaled according to F / S. This is accomplished by determining 1708 whether F / S is less than 0.6. If F / S is less than 0.6, the peripheral boost factor indicator SBF is SBF2 = SBF2 ^* (S / I ^* 1.8) (Equation 68)
It can be scaled as well. However, if it is determined 1708 that F / S is not smaller than 0.6, it can be determined 1712 whether F / S is larger than 1.8. If F / S is greater than 1.8, the second peripheral boost factor is SBF2 = SBF2 / (S / I ^* 0.6) (Equation 69)
Can be scaled 1714 as

  If the second peripheral boost factor is scaled 1710 or 1714, or if it is determined 1712 that the F / S is not greater than 1.8, then it is determined 1716 whether the F / S is greater than 1.3. be able to. If it is determined 1716 that F / S is greater than 1.3, the second peripheral boost factor can be scaled 1718 to 1.3. If the second peripheral boost factor has been scaled 1718 or if it is determined that F / S is not greater than 1.3, then it can be determined 1720 whether F / S is greater than 1.

If it is determined 1720 that F / S is greater than 1, the second peripheral mixing factors ms and mi are set to ms = ms ^* SBF2 (Equation 70)
mi = mi ^* SBF2 (Equation 71)
As set 1722 and 1724.

  If the peripheral mixing factors ms and mi are set 1722 and 1724, or if it is determined 1720 that F / S is not greater than 1, then flow returns to input channel information reception 1100 as discussed with respect to FIG. Can follow.

  Although the adjustment / modification to the mixing factor has been described as being performed after generating the mixing factor that can be utilized in the downmixer to substantially maintain the energy and intended direction of the input signal in the output signal, FIG. It will be apparent that the mixing factor adjustment discussed with respect to ˜16 may be independent of the mixing factor generation described with respect to FIGS. 4-6 and / or FIGS. Furthermore, the mixing coefficient adjustment performed in connection with FIGS. 14 to 17 can be performed at specific intervals, for example, every 64 samples of audio signal information processed by the downmixer, in this case, for example, The total sampling rate of the input signal is 44,100 samples per second. Other specific periods may be used to adjust / change the mixing factor. Further, the downmixer may be able to process the audio signal at a sampling rate other than 44,100 samples per second.

  The downmixers 100 and 900 are described above, although they are described as downmixers that downmix an input signal having three input channels or 5.1 input channels, respectively, into an output signal having two output channels. It will be apparent that the present teachings can be applied to downmixers for mixing input signals having any number of input channels into output signals having fewer output channels than the number of input channels. Downmixers 100 and 900 may be implemented with one or more microprocessors executing appropriate programmed code stored in the microprocessor's internal memory and / or storage devices 120 and 942, respectively. For example, the microprocessor is sufficiently programmed to suit the processing capabilities and other hardware requirements to allow the microprocessor to provide the functionality described herein with respect to the downmixers 100 and 900. Specific processing power and other hardware requirements. In addition, the microprocessor may be able to provide any digital signal processing function, filtering function, or other function that performs the downmixing described herein.

  While the downmixer 100 or 900 is in operation, the test mixer can be used at any time to generate the mixing coefficient value. The controller may constantly monitor input and output energy using a test mixer, such as test mixer 104 or test mixer 950, for example, and substantially maintain the signal energy and intended direction of the input signal in the output signal. Thus, one or more mixing coefficient values can be determined in a timely manner. Alternatively, the controller 106 monitors the input signal and output signal energy in the full bandwidth downmixer and causes the test downmixer to generate a mixing coefficient value when the full bandwidth output energy is not equal to the full bandwidth input energy. Can be activated.

  While the previous channel and peripheral channel mixing factor values have been described as being generated using test mixers such as test mixer 104 and preliminary test mixer 950, respectively, while the down mixer is downmixing the input signal to the output signal It will be apparent that the mixing coefficient value can be determined using a full bandwidth downmixer. In this case, the test mixer is not necessary and the test mixer may not be provided. For example, the controller 106 determines the input energy and the full bandwidth output signal of the full bandwidth input of the full bandwidth downmixer, and the limited bandwidth energy was previously described in connection with FIGS. 4-6 and 11-13. This total bandwidth energy can be used to generate and / or update the mixing factor value in a similar manner. Further, although the test downmixer 950 is described as being used with 5.1 to channel downmixer, it produces a peripheral mixing factor value that can be used in any downmixer with peripheral channel downmixing capabilities. It will be apparent that a test downmixer 950 can be utilized for this purpose.

  A downmixer is provided that can generate the mixing factor such that the energy of the input signal and the desired direction are substantially maintained in the output signal. Such mixing factor generation can be achieved with a test downmixer that can update the value of the mixing factor for a non-testing downmixer such as, for example, a full bandwidth downmixer. The test downmixer can operate on limited bandwidth input channel information so that a mixing factor value can be generated that enhances substantially audible frequencies that can be perceived by the listener. Further, the downmixer may be able to adjust the mixing coefficient value according to the ratio of energy in a certain combination of input channels (ie, the ratio of the previous channel energy to the subsequent channel energy, etc.). For example, when the input signal is downmixed, the mixing coefficient may be adjusted to emphasize the detected start of an acoustic event such as a sound from a musical instrument or a syllable of a speech. Additionally or alternatively, the mixing factor value can be adjusted to provide a more accurate interpretation of the reverberation of the input signal in the output signal. Furthermore, the downmixer may be able to maintain the intended direction of the input signal when the downmixed signal is later upmixed, for example by a decoder. The decoder may determine that peripheral channel information that is downmixed according to at least some of the teachings described herein is peripheral channel information that can be upmixed as peripheral channel information.

  Downmixers 100 and 900 are typically implemented as programming executed on one or more microprocessors to perform the functions described herein. However, it will be apparent that the downmixer can be implemented using any combination of hardware devices and / or programming executed on one or more microprocessors to perform the functions described herein.

  Similarly, the controllers 106 and 940 may be composed of any combination of hardware devices that are designed for specific functions (eg, including application specific integrated circuits that may provide functions such as filtering, mixing, etc.). . Controllers 106 and 940 may comprise a microprocessor that executes programmed code to accomplish the functions described with respect to controllers 106 and 940.

  Storage device 120 and storage device 942 may be comprised of one or more fixed or removable storage devices including but not limited to solid media, magnetic and optical media. Solid media may include integrated circuits such as ROM, PROM, EPROM, EEPROM, and any type of RAM, removable storage devices such as flash media cards, and any derivative memory system of these devices. However, it is not limited to these. Magnetic media may include, but is not limited to, magnetic tape and magnetic disks such as floppy disks and hard disk drives. Optical media may include, but is not limited to, optical discs such as compact discs and digital video discs. In general, storage devices 120 and 942 include a working memory (RAM) portion and a program memory portion that stores programmed code for a microprocessor that performs the functions described herein. Further, the storage devices 120 and 942 may include sufficient storage media to store a mixing coefficient table, for example, for downmixing the input signal described above to an output signal.

  Although the downmixers 100 and 900, and particularly the controllers 106 and 940, have been described as averaging input and output signal energy over a particular period, such as a first period, for example, the downmixers 100 and 900 may achieve averaging over other periods. It will be clear that it is good. Further, it will be apparent that at least some of the above advantages may be achieved when the input and / or output signal energy is not averaged.

  Further, although it has been described that one or more mixing factors are generated in the test mixer, it is clear that a test mixer may not be provided, in which case the full bandwidth downmixers 102 and 901 are in operation. , Each full bandwidth downmixer can generate and / or adjust the mixing factor while downmixing the input signal to the output signal to achieve at least some of the above advantages.

  While various embodiments of the present invention have been described, it will be apparent to those skilled in the art that many other embodiments and implementations are possible within the scope of the present invention. Accordingly, the invention is not limited except as in terms of the appended claims and their equivalents.

  The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily drawn to scale, emphasis is placed on illustrating the principles of the invention. Moreover, in the drawings, like reference numerals designate corresponding parts throughout the drawings.

FIG. 1 is a functional block diagram of a downmixer device for downmixing a 3-channel input signal into a 2-channel output signal. FIG. 2 is a flowchart showing the operation of the downmixer apparatus of FIG. FIG. 3 is a flowchart showing generation of a mixing coefficient between the down mixer of FIG. 1 and the down mixer of FIG. FIG. 4 is a flowchart illustrating the channel energy determination of FIG. 3 that may be used when downmixing a 3 channel input signal to a 2 channel output signal. FIG. 5 is a flowchart illustrating the determination of the feedback constant of FIG. 3 that may be used when downmixing a 3 channel input signal to a 2 channel output signal. FIG. 6 is a flowchart illustrating the generation of the channel mixing factor of FIG. 3 that can be used when downmixing a 3-channel input signal to a 2-channel output signal. FIG. 7 is a graph of mixing coefficients generated according to the flowcharts of FIGS. 4-6 for a single input signal that is panned from the center toward the left channel. FIG. 8 is a graph of the mixing factor as a function of panning angle, experimentally derived to compensate for the fine error of localization when a 3 channel signal is downmixed and reproduced through 2 channels. . FIG. 9 is a functional block diagram of a downmixer device for downmixing a 5.1 channel input signal into a 2 channel output signal. FIG. 10 is a flowchart showing the operation of the downmixer apparatus of FIG. FIG. 11 is a flowchart illustrating the determination of I / P and O / P channel energy for the generation of FIG. 3 for the downmixer of FIG. 12 is a flowchart illustrating the generation of at least one feedback constant of FIG. 3 for the downmixer of FIG. 13 is a flowchart illustrating the generation of one or more mixing coefficients of FIG. 3 for the downmixer of FIG. FIG. 14 is a flowchart illustrating adjustment of the mixing coefficient generated for the downmixer of FIG. FIG. 15 is a flowchart illustrating the channel energy determination of FIG. FIG. 16 is a flowchart illustrating adjustment of one or more mixing coefficients of FIG. FIG. 17 is a flowchart illustrating adjustment of one or more mixing coefficients of FIG.

Claims (25)

A method for establishing the coefficients for downmixing an output signal having a plurality of output channels multichannel input signal having a plurality of input channels, the method comprising,
And determining an input energy at the plurality of input channels of the input signal,
And determining the output energy to at least one of the plurality of output channels of the output signal,
Determining a first coefficient which is used to on the basis of the said input energy and the output energy, mixing the plurality of input channels over a first frequency band,
Based on the coefficients of the first, and determining a second coefficient which is used to mix the plurality of input channels for a second frequency band wider than the frequency band of the first method. Based on said input energy and the output energy, further comprises determining a feedback constant,
It said determining a second coefficient, based on said with the feedback constant first coefficient includes determining the second coefficient The method of claim 1. Wherein determining the input energy,
Based on the left channel input information and center-channel input information from the input signal, and determining a left channel input energy,
Based on the right channel input information and the center channel input information from the input signal, and a determining a right channel input energy,
Determining the output energy,
Based on the left channel output information from the output signal, and determining a left channel output energy,
Based on the right channel output information from the output signal, see contains and determining a right channel output energy,
Wherein determining the first coefficient comprises determining a left channel coefficients and right channel coefficients, A method according to claim 1 or 2. Wherein determining the input energy,
Based on the left surround input information from the input signal, determining a left surround input energy,
Based on the right surround input information from the input signal comprises determining a right surround input energy,
Determining the output energy,
Based on the left surround output information, and determining a left surround output energy,
Based on the right surround output information includes determining a right surround output energy,
It includes determining a left surround coefficients and right surround coefficients A method according to claim 1 or 2 or 3 for determining the first coefficient. Wherein before determining the input energy, the method described in further comprising, according to claim 1 or 2 or 3 or 4 to limit the bandwidth of the input signal in order to keep the audio frequency selected. The selected audio frequency is an intermediate frequency, The method of claim 5. Further comprising the method of claim 5 or 6 using the first coefficient in the test downmixer for mixing an input signal the bandwidth obtained from the multi-channel input signal is limited. It said second determining the coefficients includes determining a left channel coefficients and right channel coefficients, A method according to claim 1. It said determining a second coefficient includes determining a left channel coefficients and right channel coefficients, and determining a left surround coefficient and right surround coefficient The method of claim 1. A system for establishing the coefficients for downmixing a multi-channel input signal into an output signal, the system comprising:
A first coefficient used to mix a plurality of input channels of the multi-channel input signal over a first predetermined frequency band; and over a second predetermined frequency band wider than the first predetermined frequency band. A memory including a second coefficient used to mix the plurality of input channels of a multi-channel input signal ;
Based on the coefficients of the second, and mixers for mixing the input signal to the output signal,
Look including a controller coupled to the said memory and said mixer,
The controller,
Determining input energy in the plurality of input channels of the input signal ;
Determining an output energy for at least one output channel of the output signal ;
And that based on the input energy and the output energy, to determine the coefficients of the first,
Determining the second coefficient based on the first coefficient ;
A system that is operable to do . Wherein the input signal is viewed contains a left input channel and the central input channel and a right input channel, the output signal includes a right test output channel and a left test output channels, the system according to claim 10. Said controller by said left input channel based on the central input channel to determine the left channel input energy, and determines the right channel input energy based and the right input channel and the central input channel, Operable to determine the input energy ;
Wherein the controller determines a left channel output energy based on the left-test output channels, and, by determining the right channel output energy based on the right test output channel, operable to determine the output energy And
Wherein the controller is by determining a left channel coefficients and right channel coefficients based on said input energy and the output energy is operable to determine said second coefficient of claim 11 system. Wherein the controller determines a left surround input energy based on the left surround input channels and by determining the right surround input energy based on the right surround input channels is operable to determine the input energy ,
Wherein the controller determines a left surround output energy based on the left surround test output channels, and, by determining the right surround output energy based on the right surround test output channel, operable to determine the output energy And
Wherein the controller is by determining a left surround coefficient and right surround coefficient based on said input energy and the output energy is operable to determine said second coefficient, claim 10 or 11, or 12. The system according to 12. Wherein the controller, prior to determining the input energy is operable to limit the bandwidth of the input signal in order to keep the audio frequency chosen, according to claim 10 or 11 or 12 or 13 system. The selected audio frequency is an intermediate frequency system according to claim 14. Said controller, said determining a second left channel coefficients based on the left channel coefficients, and by determining the second right channel coefficients based on the right channel coefficients, to determine a second coefficient The system of claim 12, wherein the system is operable . Wherein the controller, the left second left surround coefficients determined on the basis of the surround factor is further operable to determine a second right surround coefficient based on said right surround factor, according to claim 16 System. A first down mixer containing the coefficients for mixing a plurality of input channels to at least one output channel over a first frequency band,
A test downmixer coupled to the first downmixer, comprising:
The test downmixer is operable to determine the test factor for a second frequency band narrower than the frequency band of the first to update the coefficient of the first down mixer, down mixer. It said plurality of input channels includes a left channel and a right channel and the center channel, down mixer according to claim 18. It said plurality of input channels includes a left surround channel and the right surround channel, down mixer according to claim 18. The first down mixer includes a multi-channel input signal input across the first frequency band, the test down mixer includes a multi-channel input signal input over the second frequency band, according to claim 18 Down mixer. The test down mixer receives the intermediate audio frequency from the multi-channel input signal existing in the multichannel input signal is input, the down mixer according to claim 18. The coefficient is determined based on the input energy and output energy, down mixer according to claim 18. The coefficients are updated during operation of the down mixer, down mixer according to claim 18. The coefficients, and a right coefficients are updated according to the left coefficients and right feedback parameters are updated according to the left feedback parameters, down mixer according to claim 24.

Images